Hasil untuk "Dictionaries and other general reference works"

Yusuke Ide, Adam Nohejl, Joshua Tanner et al.

We study dictionary definition generation (DDG), i.e., the generation of non-contextualized definitions for given headwords. Dictionary definitions are an essential resource for learning word senses, but manually creating them is costly, which motivates us to automate the process. Specifically, we address learner's dictionary definition generation (LDDG), where definitions should consist of simple words. First, we introduce a reliable evaluation approach for DDG, based on our new evaluation criteria and powered by an LLM-as-a-judge. To provide reference definitions for the evaluation, we also construct a Japanese dataset in collaboration with a professional lexicographer. Validation results demonstrate that our evaluation approach agrees reasonably well with human annotators. Second, we propose an LDDG approach via iterative simplification with an LLM. Experimental results indicate that definitions generated by our approach achieve high scores on our criteria while maintaining lexical simplicity.

Peng Zheng, Dehong Gao, Deng-Ping Fan et al.

We introduce a novel bilateral reference framework (BiRefNet) for high-resolution dichotomous image segmentation (DIS). It comprises two essential components: the localization module (LM) and the reconstruction module (RM) with our proposed bilateral reference (BiRef). The LM aids in object localization using global semantic information. Within the RM, we utilize BiRef for the reconstruction process, where hierarchical patches of images provide the source reference and gradient maps serve as the target reference. These components collaborate to generate the final predicted maps. We also introduce auxiliary gradient supervision to enhance focus on regions with finer details. Furthermore, we outline practical training strategies tailored for DIS to improve map quality and training process. To validate the general applicability of our approach, we conduct extensive experiments on four tasks to evince that BiRefNet exhibits remarkable performance, outperforming task-specific cutting-edge methods across all benchmarks. Our codes are available at https://github.com/ZhengPeng7/BiRefNet.

F. L. Carneiro, S. C. Ulhoa, M. P. Lobo

This article explores the differences between frame and coordinate transformations in relativistic theories. We highlight the key role of tetrad fields in connecting spacetime and frame indices. Using Maxwell's electrodynamics as an example, we show that Maxwell's equations are invariant under coordinate transformations but exhibit covariant behavior under frame transformations. We also analyze the energy-momentum of an electromagnetic field in different frames, providing deeper insights into the implications of different frames of reference and coordinate systems.

Halim Cagri Ates, Shruti Bhargava, Site Li et al.

Successfully handling context is essential for any dialog understanding task. This context maybe be conversational (relying on previous user queries or system responses), visual (relying on what the user sees, for example, on their screen), or background (based on signals such as a ringing alarm or playing music). In this work, we present an overview of MARRS, or Multimodal Reference Resolution System, an on-device framework within a Natural Language Understanding system, responsible for handling conversational, visual and background context. In particular, we present different machine learning models to enable handing contextual queries; specifically, one to enable reference resolution, and one to handle context via query rewriting. We also describe how these models complement each other to form a unified, coherent, lightweight system that can understand context while preserving user privacy.

Nicola Bamonti

This work introduces a novel three-fold classification of reference frames in General Relativity, distinguishing between Idealised Reference Frames (IRFs), Dynamical Reference Frames (DRFs), and Real Reference Frames (RRFs). By defining a reference frame as a set of degrees of freedom instantiated by a physical system, the work contrasts this notion with that of coordinate systems-purely mathematical idealisations lacking physical instantiation. This classification addresses two longstanding challenges in GR: (P1) the difficulty of defining local and gauge-invariant observables, and (P2) how to interpret diffeomorphism gauge freedom in physical terms rather than as merely a mathematical redundancy. Overall, this work clarifies the conceptual foundations in classical General Relativity, enhancing our understanding of gauge-symmetries, observers and laying the groundwork for future investigations in both classical and quantum gravitational contexts.

Ashkan Mirzaei, Tristan Aumentado-Armstrong, Marcus A. Brubaker et al.

The popularity of Neural Radiance Fields (NeRFs) for view synthesis has led to a desire for NeRF editing tools. Here, we focus on inpainting regions in a view-consistent and controllable manner. In addition to the typical NeRF inputs and masks delineating the unwanted region in each view, we require only a single inpainted view of the scene, i.e., a reference view. We use monocular depth estimators to back-project the inpainted view to the correct 3D positions. Then, via a novel rendering technique, a bilateral solver can construct view-dependent effects in non-reference views, making the inpainted region appear consistent from any view. For non-reference disoccluded regions, which cannot be supervised by the single reference view, we devise a method based on image inpainters to guide both the geometry and appearance. Our approach shows superior performance to NeRF inpainting baselines, with the additional advantage that a user can control the generated scene via a single inpainted image. Project page: https://ashmrz.github.io/reference-guided-3d

Margrét Jónsdóttir

This article deals with the idiom sjá sig eftir, a variant of sjá eftir, more or less in the same meaning as iðrast ‛repent, regret’. The experiencer subject is in the nominative. The accusative form of the reflexive pronoun, sig, theorectically the indirect object of the verb, is bound and commanded by the next subject. Consequently, the binding domain is short and the pronoun is a short-distance one. Semantically, it could be argued that the reflexive pronoun is a part of the verb which is inherently reflexive. Syntactically, the reflexive pronoun in question behaves mostly like other obligatory reflexive pronouns. The examples of sjá sig eftir, from written texts as well as the spoken language, are from the middle of the nineteenth century to the last quarter of the twentieth. Most of the examples are from the eastern part of Iceland, and this includes all of the spoken language examples. It could be argued that the examples of sjá sig eftir are relics. It doesn‘t have to be so, as there are sporadic examples of other experiencer-subject verbs, normally not reflexive, as reflexive ones (e.g. hlakka (sig) til ‘look forward to’). Some of the examples are from child language. Furthermore, there are corresponding examples from related languages that show the same behaviour.

Vladimir Ivanov, Valery Solovyev

Concrete/abstract words are used in a growing number of psychological and neurophysiological research. For a few languages, large dictionaries have been created manually. This is a very time-consuming and costly process. To generate large high-quality dictionaries of concrete/abstract words automatically one needs extrapolating the expert assessments obtained on smaller samples. The research question that arises is how small such samples should be to do a good enough extrapolation. In this paper, we present a method for automatic ranking concreteness of words and propose an approach to significantly decrease amount of expert assessment. The method has been evaluated on a large test set for English. The quality of the constructed dictionaries is comparable to the expert ones. The correlation between predicted and expert ratings is higher comparing to the state-of-the-art methods.

Amir Shaikhha, Mahdi Ghorbani, Hesam Shahrokhi

This article introduces hinted dictionaries for expressing efficient ordered sets and maps functionally. As opposed to the traditional ordered dictionaries with logarithmic operations, hinted dictionaries can achieve better performance by using cursor-like objects referred to as hints. Hinted dictionaries unify the interfaces of imperative ordered dictionaries (e.g., C++ maps) and functional ones (e.g., Adams' sets). We show that such dictionaries can use sorted arrays, unbalanced trees, and balanced trees as their underlying representations. Throughout the article, we use Scala to present the different components of hinted dictionaries. We also provide a C++ implementation to evaluate the effectiveness of hinted dictionaries. Hinted dictionaries provide superior performance for set-set operations in comparison with the standard library of C++. Also, they show a competitive performance in comparison with the SciPy library for sparse vector operations.

Yang Li, Xiaoxue Chen, Hao Zhao et al.

We study embodied reference understanding, the task of locating referents using embodied gestural signals and language references. Human studies have revealed that objects referred to or pointed to do not lie on the elbow-wrist line, a common misconception; instead, they lie on the so-called virtual touch line. However, existing human pose representations fail to incorporate the virtual touch line. To tackle this problem, we devise the touch-line transformer: It takes as input tokenized visual and textual features and simultaneously predicts the referent's bounding box and a touch-line vector. Leveraging this touch-line prior, we further devise a geometric consistency loss that encourages the co-linearity between referents and touch lines. Using the touch-line as gestural information improves model performances significantly. Experiments on the YouRefIt dataset show our method achieves a +25.0% accuracy improvement under the 0.75 IoU criterion, closing 63.6% of the gap between model and human performances. Furthermore, we computationally verify prior human studies by showing that computational models more accurately locate referents when using the virtual touch line than when using the elbow-wrist line.

Cunliang Kong, Xuezhi Fang, Liner Yang et al.

Dictionaries can help language learners to learn vocabulary by providing definitions of words. Since traditional dictionaries present word senses as discrete items in predefined inventories, they fall short of flexibility, which is required in providing specific meanings of words in particular contexts. In this paper, we introduce the LitMind Dictionary (https://dictionary.litmind.ink), an open-source online generative dictionary that takes a word and context containing the word as input and automatically generates a definition as output. Incorporating state-of-the-art definition generation models, it supports not only Chinese and English, but also Chinese-English cross-lingual queries. Moreover, it has a user-friendly front-end design that can help users understand the query words quickly and easily. All the code and data are available at https://github.com/blcuicall/litmind-dictionary.

Stephen Ellis, Shuofei Zhu, Nobuko Yoshida et al.

Go is a popular statically-typed industrial programming language. To aid the type safe reuse of code, the recent Go release (Go 1.18) published on 15th March 2022 includes bounded parametric polymorphism via generic types. Go 1.18 implements generic types using combination of monomorphisation and call-graph based dictionary-passing called hybrid. This hybrid approach can be viewed as an optimised form of monomorphisation that statically generates specialised methods and types based on possible instantiations. A monolithic dictionary supplements information lost during monomorphisation, and it is structured according to the program's call graph. Unfortunately, the hybrid approach still suffers from code bloat, poor compilation speed, and limited code coverage. In this paper we propose and formalise a new non-specialising call-site based dictionary-passing translation. Our call-site based translation creates individual dictionaries for each type parameter, with dictionary construction occurring in place of instantiation, overcoming the limitations of hybrid. We prove it correct using a novel and general bisimulation up to technique. To better understand how different generics translations approaches work in practice, we benchmark five translators, Go 1.18, two existing monomorphisation translators, our dictionary-passing translator, and erasure translator. Our findings reveal several suggestions for improvements for Go 1.18 -- specifically how to overcome the expressiveness limitations of generic Go, and improve compile time and compiled code size performance of Go 1.18.

Ágústa Þorbergsdóttir

This paper describes the results of a survey conducted in 2020 about language issues in the financial sector. The aim of the survey was to collect information about practices concerning the use of terminology: whether it is important in the field, whether terminology is registered and collected in a systematic way, and how willing the financial institutions are to share their termbanks. The results show that there is clear interest among participants in having access to a bank of financial terminology, but slightly less so in participating in information-sharing with others, and still less so in devoting time and manpower of their own to create collections. Also, even though majority of respondents estimated that their staff look up terminology on a weekly basis (and sometimes more often), most participants indicated that their organization did not keep a register of financial terminology. At the same time, a majority believed the most pressing issue related to terminology in the workplace was that up-to-date terms had not been collected and made available to all. This lack of availability of Icelandic- language collections, both for up-to-date terms and in general, was reinforced when the organizations were asked where their staff would look for assistance with translations of financial terminology. The results emphasize the need for a good, upto- date termbank and the willingness to use terms in Icelandic.

Þorsteinn G. Indriðason

The article discusses suffix combinations in Icelandic with nominal and adjectival suffixes as first members of the combinations, how common these combinations are and which selectional restrictions are central in Icelandic derivational morphology. Combinations of twenty-six nominal suffixes and twenty-two suffixes that can attach to nominal suffixes were studied, in addition to combinations of nine adjectival suffixes and twelve other suffixes that can attach to adjectival suffixes: a total of 661 possible combinations when combinations of identical suffixes had been excluded. Of these 661 possible combinations, 36 were confirmed, using the corpora Íslenskur orðasjóður and Beygingarlýsing íslensks nútímamáls for documentation. Possible reasons for the relatively small number of confirmed suffix combinations are discussed. Selectional restrictions play a major role in explaining this. For instance, some suffixes can not attach to base words that are themselves suffixed, although they otherwise can attach to base words of two syllables. Furthermore, it turns out that it is quite difficult to change the order of suffixes in a combination. These two restrictions significantly reduce the number of possible combinations in addition to some closed suffixes that do not allow further suffixation. In Icelandic, the so called ‚split suffixation‘ exists that partially compensates for the lack of suffix combinations. The split suffixation occurs where a linking element is inserted between two suffixes that can not in many instances attach to each other. The linking element then opens for further suffixation.

Margrét Jónsdóttir

This article deals with Icelandic polysyllabic nouns of Greek-Latin origin that end in –ík, for example epík ‘epic’, keramík ‘ceramic’, and klassík ‘classic’. The oldest loan words were borrowed from Danish via German, but in recent decades they have most likely entered Icelandic through English. The main research question in this article concerns declension: how are these nouns declined in the genitive form? As feminine nouns, the words in question follow the same pattern as the monosyllabic, feminine noun vík. Thus, as vík, polysyllabic -ík nouns have the ending -ur in genitive singular. However, the ending -ar is also possible in many instances. In this paper, it is argued that the -ík ending has triggered the gender and consequently the declension of the words in question. Finally, the paper addresses the question whether -ík should be understood as an ending or as a part of the stem. As the ending -ík is not productive in the formation of new Icelandic words it is concluded that the latter is more likely.

Olga Karpova

The article is devoted to the description of new trends in theory and dictionary making process of modern English lexicography. At the same time the paper also covers the main historic steps of formation and development of national English lexicography with special reference to the most reliable English dictionaries for general purposes (early glossaries and concordances, Samuel Johnson’s Dictionary, Oxford English Dictionary, etc.) and special purposes (English writers’ glossaries, concordances, lexicons to the complete and separate works of Chaucer, Shakespeare, and other famous English men of letters). The main accent is made on the digital époque of English national lexicography, describing innovative features of both printed and Internet dictionaries of various types and formats from the point of view of a user studying English as a foreign language. The paper touches upon new branches of English lexicography (collaborative, volunteer) with users’ needs and demands at the centre of dictionary making process.

Yuki Minamisawa

This article analyzes the conceptual metaphors of ANGER in Icelandic, rendered as REIÐI. Since Lakoff & Johnson (1980), many studies have illustrated how we conceptualize emotions. Much research has been devoted to ANGER in particular, and it has been found that the conceptual metaphor ANGER IS A HOT FLUID IN A CONTAINER is present in many different languages. When it comes to Icelandic, little discussion has been devoted to such metaphors. Therefore, the present paper aims to examine what metaphors are central to the emotion of ANGER. In determining the centrality, this study uses the MI score. The results show that the central metaphor for REIÐI is ANGER IS A HOT FLUID IN A CONTAINER. However, the data also show that this metaphor highlights different aspects of the emotion, and that other conceptual metaphors such as ANGER IS FIRE and ANGER IS A DANGEROUS ANIMAL are not as strongly connected to the emotion in Icelandic as in English.

Pranjal Awasthi, Aravindan Vijayaraghavan

Dictionary learning is a popular approach for inferring a hidden basis or dictionary in which data has a sparse representation. Data generated from the dictionary A (an n by m matrix, with m > n in the over-complete setting) is given by Y = AX where X is a matrix whose columns have supports chosen from a distribution over k-sparse vectors, and the non-zero values chosen from a symmetric distribution. Given Y, the goal is to recover A and X in polynomial time. Existing algorithms give polytime guarantees for recovering incoherent dictionaries, under strong distributional assumptions both on the supports of the columns of X, and on the values of the non-zero entries. In this work, we study the following question: Can we design efficient algorithms for recovering dictionaries when the supports of the columns of X are arbitrary? To address this question while circumventing the issue of non-identifiability, we study a natural semirandom model for dictionary learning where there are a large number of samples $y=Ax$ with arbitrary k-sparse supports for x, along with a few samples where the sparse supports are chosen uniformly at random. While the few samples with random supports ensures identifiability, the support distribution can look almost arbitrary in aggregate. Hence existing algorithmic techniques seem to break down as they make strong assumptions on the supports. Our main contribution is a new polynomial time algorithm for learning incoherent over-complete dictionaries that works under the semirandom model. Additionally the same algorithm provides polynomial time guarantees in new parameter regimes when the supports are fully random. Finally using these techniques, we also identify a minimal set of conditions on the supports under which the dictionary can be (information theoretically) recovered from polynomial samples for almost linear sparsity, i.e., $k=\tilde{O}(n)$.

Yu Bai, Qijia Jiang, Ju Sun

This paper concerns dictionary learning, i.e., sparse coding, a fundamental representation learning problem. We show that a subgradient descent algorithm, with random initialization, can provably recover orthogonal dictionaries on a natural nonsmooth, nonconvex $\ell_1$ minimization formulation of the problem, under mild statistical assumptions on the data. This is in contrast to previous provable methods that require either expensive computation or delicate initialization schemes. Our analysis develops several tools for characterizing landscapes of nonsmooth functions, which might be of independent interest for provable training of deep networks with nonsmooth activations (e.g., ReLU), among numerous other applications. Preliminary experiments corroborate our analysis and show that our algorithm works well empirically in recovering orthogonal dictionaries.

M. J. Luo

This paper is a generalization of earlier papers [Nucl. Phys. B 884, 344 (2014) (arXiv:1312.2759) and JHEP 6, 63 (2015) (arXiv:1401.2488)]. We generalize the idea of quantum clock time to quantum spacetime reference frame via physical realization of a reference system by quantum rulers and clocks. Omitting the internal degrees of freedom (such as spins) of the physical rulers and clocks, only considering their metric properties, the spacetime reference frame is described by a bosonic non-linear sigma model (NLSM). We study the quantum behavior of the system under approximations, and obtain (1) a cosmological constant valued $(2/π)ρ_{c0}$ ($ρ_{c0}$ the critical density at near current epoch) which is very close to the observations; (2) an effective Einstein-Hilbert term; (3) the ratio of variance to mean-squared of spacetime interval tends to a universal constant $2/π$ in the infrared region. This effect is testable by observing a linear dependence between the inherent quantum variance and mean-squared of the redshifts from cosmic distant spectral lines. The proportionality is expected to be the observed percentage of the dark energy. The equivalence principle is also generalized to the quantum level.